Court Further Refines Search Protocol, Adds Search Terms and Orders Distinct Conjunctive and Disjunctive Keyword Searches
ClearOne Communications, Inc. v. Chiang, 2008 WL 920336 (D. Utah Apr. 1, 2008)
This decision further refines the search protocol to be used to search data from computers used by certain defendants, which were imaged pursuant to two court orders issued in 2007. The second imaging order sought to establish a protocol for searching the mirror images to identify relevant and responsive documents. That protocol was refined in a November 5, 2007 order. The protocol in place required keyword searches by technical experts; review of search result reports by defense counsel for facial claims of privilege; delivery of the reports to plaintiff’s counsel for preliminary assertion of responsiveness; defense counsel's review for responsiveness and privilege; and delivery of documents and privilege logs.
The parties had agreed on many search terms. Plaintiff had accepted, with five additions, the search terms proposed by defendants in September 2007. The issue decided on this motion related to the five additional search terms proposed by plaintiff, and the conjunctive or disjunctive use of the terms in the searches to be conducted.
The court first analyzed whether a conjunctive or disjunctive search should be used:
“Essentially, there are three categories of search criteria: ‘Name’ (searching for names of specific individuals); ‘Tech’ (searching for a particular technological reference); and ‘License’ terms (searching for terms relating to the licensing of certain source code).” WideBand Defendants say it is “reasonable to require some connectors that would narrow the search results to subjects relevant to the issues in this lawsuit.” Specifically, WideBand Defendants request “that the ‘Name’ and ‘License’ search terms be combined with the ‘Tech’ terms.”
As to the “Name” terms, conjunctive search seems necessary. Otherwise, every occurrence of the “Name” terms will result in a positive hit, meaning that virtually every document in the electronic media will be identified as potentially responsive. In a relatively small business such as WideBand, almost every document will refer to one of the key employees in the company. Requiring a hit of one “Name” term AND one “Tech” term will ensure that more responsive documents are flagged as potentially responsive.
However, as to the “License” terms, conjunctive search could be excessively narrow. Again, because WideBand is a relatively small company, licensing activity would be relatively small. By comparison, technology is the core of its business, so disjunctive use of the “Tech” terms would probably result in an excessive number of false positives.
(Footnotes and citations to the record omitted.) Thus, the court ruled that searches were to be conducted (1) for the occurrence of “License” terms, and (2) for the concurrent occurrence of “Name” AND “Tech” terms.
The court also found that the five additional search terms proposed by plaintiff were not extremely broad, and that the use of the terms, in the disjunctive, would yield evidence that was potentially very significant to the case. Thus, it ruled that the list of terms already agreed to by the parties, together with the five additional terms requested by plaintiff, should be used.
The court cautioned that the search protocol was just one step in conducting e-discovery in the case:
[T]he search protocol is not the “last word” on electronic discovery in this case. The use of key word protocols is one step in the process which contemplates many more steps, including review of search result reports by WideBand Defendants' counsel for facial claims of privilege; delivery of reports to ClearOne counsel for preliminary assertion of responsiveness; WideBand Defendants' counsel's review of the reports and documents for responsiveness and privilege; and delivery of documents and privilege logs.
This order is not the last word on key words, either. If documents are discovered which suggest that other documents exist which were not identified as potentially responsive, or if a surprisingly small or unreasonably large number of documents is identified as potentially responsive, refinement may be needed. Much of the argument is now speculative, since there is no actual experience with a search. This first protocol may suffice, or it may in effect be a sampling which reveals the need for more-or less-or different-key words.